PathFinder

AI-powered job-to-roadmap generator that transforms skill gaps into actionable 30-day learning plans

Inspiration

As students preparing for internships and full-time roles, we repeatedly faced the same problem: we could see job descriptions listing dozens of required skills, but we didn't know exactly where we stood or how to close the gap efficiently.

Resumes and job descriptions exist separately, and manually comparing them is time-consuming and often inaccurate. Even after identifying missing skills, creating a structured, realistic learning plan is another challenge.

PathFinder was inspired by this gap between awareness and action. We wanted to build a system that doesn't just analyze skills - but transforms analysis into a concrete, personalized 30-day roadmap.

Features

PathFinder is an AI-powered job-to-roadmap generator that:

  • Resume Analysis - Supports PDF, TXT, DOC, DOCX formats
  • Skill Extraction - Automatically extracts skills from resumes and job descriptions
  • Match Calculation - Computes accurate skill match percentage
  • Gap Identification - Identifies missing skills between resume and job requirements
  • 30-Day Roadmap - Generates structured, day-by-day learning plans
  • Personalization - Customizes roadmaps based on:
    • Daily time commitment
    • Experience level
    • Preferred study schedule
  • Roadmap Comparison - Compare default vs personalized learning paths
  • Progress Tracking - Track completion and maintain learning streaks

Instead of telling users "You are missing React," it tells them exactly what to do each day to learn it.

🛠️ Tech Stack

Frontend

  • React 19 - Component-based UI
  • Next.js 16 (App Router) - Routing and API integration
  • Tailwind CSS - Responsive design
  • Framer Motion - Interactive animations

Backend

  • Next.js API Routes - Resume-job analysis
  • Node.js - Server-side logic
  • OpenAI API - Natural language processing

🔧 How It Works

Core Workflow

  1. Resume Upload → PDF.js extracts raw text
  2. Skill Extraction → OpenAI identifies categorized skills
  3. Matching Algorithm → Compare resume skills vs job skills
  4. Score Calculation → Compute match percentage
  5. Roadmap Generation → Distribute missing skills across 30 days

Skill Prioritization Algorithm

Skill prioritization follows a weighted approach:

Priority_i = Skill Importance_i / Estimated Learning Time_i

This ensures high-impact skills are scheduled earlier while balancing workload.

💪 Challenges We Overcame

1. Resume Parsing Complexity

PDF resumes often contain multi-column layouts, tables, and formatting inconsistencies. Extracting clean text required normalization and fallback handling.

2. AI Prompt Engineering

Job descriptions vary greatly in structure. Fine-tuning prompts to reliably extract accurate skill lists without hallucinations required iterative testing.

3. Personalization Logic

Designing a roadmap that adapts to:

  • Different daily time limits
  • Beginner vs intermediate learners
  • Weekday vs weekend schedules

required careful distribution algorithms.

4. Development Environment Issues

We encountered cache locking and persistence errors during development due to Turbopack and file synchronization conflicts. Debugging these issues improved our understanding of build systems and local environments.

Accomplishments

  • Successfully integrated AI into a real-world career use case
  • Built a complete end-to-end pipeline: upload → analyze → roadmap
  • Designed a clean, intuitive UI with interactive roadmap comparison
  • Implemented personalized scheduling logic rather than static templates
  • Created a tool that provides actionable outputs, not just analysis

Most importantly, we transformed a common frustration into a structured solution.

📚 What We Learned

  • How to integrate AI APIs into production-ready web applications
  • How to design full-stack systems using Next.js App Router
  • The importance of prompt engineering for reliable AI outputs
  • How to build user-centric features instead of purely technical features
  • Debugging build tools, caching systems, and environment configurations
  • Translating abstract ideas (like "skill gaps") into algorithmic logic

We also learned that building impactful tools requires both engineering precision and empathy for the user.

What's Next for PathFinder

We plan to expand PathFinder with:

  • [ ] Real-time skill validation quizzes
  • [ ] Integration with online learning platforms
  • [ ] AI-generated resource recommendations (courses, videos, documentation)
  • [ ] Progress analytics dashboard
  • [ ] Peer comparison and community roadmap sharing
  • [ ] Mobile-friendly and PWA support
  • [ ] Multi-language support

Our long-term vision is to evolve PathFinder into a full AI-powered career acceleration platform.

Built with ❤️ for students navigating the job search journey

Built With

  • css
  • framer
  • next.js
  • openai
  • pdf.js
  • tailwind
Share this project:

Updates